NOT BORN YESTERDAY
NOT
BORN
YESTERDAY
the science of Who we trust
and What we believe
hugo mercier
pr ince ton university pr ess
pr ince ton & oxfor d
Copyright © 2020 by Hugo Mercier
Requests for permission to reproduce material from this work
should be sent to permissions@press.princeton.edu
Published by Prince ton University Press
41 Wil iam Street, Prince ton, New Jersey 08540
6 Oxford Street, Woodstock, Oxfordshire OX20 1TR
press .princeton .edu
All Rights Reserved
ISBN 978-0-691-17870-7
ISBN (e- book) 978-0-691-19884-2
British Library Cataloging- in- Publication Data is available
Editorial: Sarah Caro, Charlie Allen, and Hannah Paul
Production Editorial: Jenny Wolkowicki
Text design: Leslie Flis
Jacket design: Michel Vrana
Production: Erin Suydam
Publicity: Maria Whelan and Kate Farquhar- Thomson
Jacket art: iStock
This book has been composed in Arno Pro and Heading Smallcase Pro
Printed on acid- free paper. ∞
Printed in the United States of Amer ica
10 9 8 7 6 5 4 3 2 1
To Thérèse Cronin
CONTENTS
List of Illustrations ix
Acknowl edgments xi
Introduction xiii
1 The Case for Gullibility
1
2 Vigilance in Communication
15
3 Evolving Open- Mindedness
30
4 What to Believe?
47
5 Who Knows Best?
63
6 Who to Trust?
78
7 What to Feel?
95
8 Demagogues, Prophets, and Preachers
113
9 Propagandists, Campaigners, and Advertisers
128
10 Titil ating Rumors
146
vii
viii con t en t s
11 From Circular Reporting to Super natural Beliefs
166
12 Witches’ Confessions and Other Useful
Absurdities 181
13 Futile Fake News
199
14 Shallow Gurus
217
15 Angry Pundits and Skil ful Con Men
240
16 The Case against Gullibility
257
Notes 273
References 307
Index 351
LIST OF ILLUSTR ATIONS
Figure 1. The lines in the Asch conformity experiments.
6
Figure 2. “Bridge” strip from the webcomic xkcd
by Randall Munroe.
72
Figure 3. Two examples of pareidolia: seeing faces
where there are none.
157
Figure 4. What path does a ball launched at the arrow
follow when it exits the tube?
224
ix
ACKNOWL EDGMENTS
the idea for this book stems from the article “Epistemic
Vigilance,” written by Dan Sperber, Fabrice Clément, Christophe
Heintz, Olivier Mascaro, Gloria Origgi, Deirdre Wilson, and my-
self. In this article we suggested that humans are endowed with
cognitive mechanisms dedicated to evaluating communicated
information. I am particularly grateful to Dan Sperber— thesis
supervisor, coauthor, mentor, friend. Besides having shaped my
ideas through his writings and his discussions, he patiently read
and gave me feedback on the book. Fabrice Clément had writ-
ten his PhD thesis and a book— Les Mécanismes de la crédulité
(The mechanisms of credulity)—on the same theme, and we
discussed these issues when I was a postdoctoral researcher at
the University of Neuchâtel. Besides Dan and Fabrice, the ideas
in this book have been shaped by the feedback from the students
of the Communication, Trust, and Argumentation class from
2018 to 2019, and by discussions at the Department of Cognitive
Studies of the ENS in Paris; the University of Pennsylvania; and
the countless conferences, restaurants, pubs, and cafés where
I’ve badgered people with the idea that humans aren’t gullible.
Lila San Roque generously shared fantastic examples of eviden-
tial use among the Duna, and I benefited from Chris Street’s
knowledge of the lie detection lit er a ture.
xi
xi ackno w l edgmen t s
My deepest thanks go to those who have commented on the
whole, or parts of the manuscript: Sacha Altay (twice!), Stefaan
Blancke, Pascal Boyer, Coralie Cheval ier, Thérèse Cronin (twice
as well!), Guil aume Dezecache, Helena Miton, Olivier Morin,
Thom Scott- Phil ips, Dan Sperber, and Radu Umbres.
This book would not have existed without my agents John and
Max Brockman, Sarah Caro, the editor who believed in the proj-
ect from the start and provided very valuable feedback, as well
as the team at Prince ton University Press.
I have benefited from the financial backing of the Direction
Générale de l’Armement (thanks to Didier Bazalgette in par tic-
u lar); the Philosophy, Politics, and Economics program at the
University of Pennsylvania (with the generous support of
Steven F. Goldstone); the University of Neuchâtel’s Cognitive
Science group and the Swiss National Science Foundation
(Ambizione grant no. PZ00P1_142388); the Agence Nationale
de la Recherche (grant EUR FrontCog ANR-17- EURE-0017
to the DEC, and grant ANR-16- TERC-0001-01 to myself); and
last but not least, the Centre National de la Recherche Scienti-
fique, my current employer, allowing me to work in the amazing
place that is the Jean Nicod Institute. In par tic u lar, the team I
belong to— Evolution and Social Cognition, composed of Jean-
Baptiste André, Nicolas Baumard, Coralie Chevallier, Olivier
Morin, and our students, engineers, and postdocs— provides the
best social and intellectual environment one could hope for.
I will never stop thanking my parents, grandparents, and ex-
tended family for their unwavering support. Christopher and
Arthur are the best boys in the world; they have taught me much
about love. They have also taught me that children aren’t gull-
ible, even when we wish they’d be a bit easier to influence.
Thérèse’s encouragement has meant more to me than I can
ever communicate. Thank you for every thing.
INTRODUCTION
as i was walking back from university one day, a
respectable- looking middle- aged man accosted me. He spun a
good story: he was a doctor working in the local hospital, he had
to rush to some urgent doctorly thing, but he’d lost his wallet,
and he had no money for a cab ride. He was in dire need of twenty
euros. He gave me his business card, told me I could call the num-
ber and his secretary would wire the money back to me shortly.
After some more cajoling I gav
e him twenty euros.
There was no doctor of this name, and no secretary at the end
of the line.
How stupid was I?
And how ironic that, twenty years later, I would be writing a
book arguing that people aren’t gullible.
The Case for Gullibility
If you think I’m gullible, wait until you meet, in the pages that
follow, people who believe that the earth is a flat disk surrounded
by a two- hundred- foot wall of ice, “Game of Thrones–style,”1 that
witches poison their cattle with magical darts, that the local Jews
kill young boys to drink their blood as a Passover ritual, that high-
up Demo cratic operatives oversee a pedophile ring out of a
xiii
xiv in t roduc t ion
pizza joint, that former North Korean leader Kim Jong-il could
teleport and control the weather, or that former U.S. president
Barack Obama is a devout Muslim.
Look at all the gibberish transmitted through TV, books,
radio, pamphlets, and social media that ends up being accepted
by large swaths of the population. How could I possibly be claim-
ing that we aren’t gullible, that we don’t accept what ever we read
or hear?
Arguing against widespread credulity puts me in the minor-
ity. A long line of scholarship— from ancient Greece to twenty-
first- century Amer ica, from the most progressive to the most
reactionary— portrays the mass of people as hopelessly gullible.
For most of history, thinkers have based their grim conclusions
on what they thought they observed: voters submissively fol ow-
ing demagogues, crowds worked up into rampages by blood-
thirsty leaders, masses cowing to charismatic personalities. In the
mid- twentieth century, psychological experiments brought more
grist to this mil , showing participants blindly obeying author-
ity, believing a group over the clear evidence of their own eyes.
In the past few de cades, a series of sophisticated models have
appeared that provide an explanation for human gullibility. Here
is the core of their argument: we have so much to learn from
others, and the task of figuring out who to learn from is so dif-
ficult, that we rely on simple heuristics such as “follow the ma-
jority” or “follow prestigious individuals.” Humans would owe
their success as a species to their capacity to absorb their local
culture, even if that means accepting some maladaptive practices
or mistaken beliefs along the way.
The goal of this book is to show this is all wrong. We don’t
credulously accept what ever we’re told— even if those views are
supported by the majority of the population, or by prestigious,
in t r o d u c t i o n xv
charismatic individuals. On the contrary, we are skilled at figur-
ing out who to trust and what to believe, and, if anything, we’re
too hard rather than too easy to influence.
The Case against Gullibility
Even if suggestibility might have some advantages in helping us
acquire skil s and beliefs from our cultural environment, it is sim-
ply too costly to be a stable, per sis tent state of affairs, as I wil
argue in chapter 2. Accepting what ever others are communicat-
ing only pays off if their interests are aligned with ours— think
cel s in a body, bees in a beehive. As far as communication be-
tween humans is concerned, such commonality of interests is
rarely achieved; even a pregnant mother has reasons to mistrust
the chemical signals sent by her fetus. Fortunately, there are ways
of making communication work even in the most adversarial of
relationships. A prey can convince a predator not to chase it. But
for such communication to occur, there must be strong guaran-
tees that those who receive the signal will be better off believing
it. The messages have to be kept, on the whole, honest. In the case
of humans, honesty is maintained by a set of cognitive mecha-
nisms that evaluate communicated information. These mecha-
nisms allow us to accept most beneficial messages—to be
open— while rejecting most harmful messages—to be vigilant.
As a result, I have called them open vigilance mechanisms, and they
are at the heart of this book.2
What about the “observations” used by so many scholars to
make the case for gullibility? Most are merely popu lar miscon-
ceptions. As the research reviewed in chapters 8 and 9 shows,
those who attempt to persuade the masses— from demagogues
to advertisers, from preachers to campaign operatives— nearly
xvi in t roduc t ion
always fail miserably. Medieval peasants in Eu rope drove many
a priest to despair with their stubborn re sis tance to Christian pre-
cepts. The net effect on presidential elections of sending flyers,
robocalling, and other campaign tricks is close to zero. The sup-
posedly all- powerful Nazi propaganda machine barely affected
its audience—it couldn’t even get the Germans to like the Nazis.
Sheer gullibility predicts that influence is easy. It is not. Stil ,
indubitably, people sometimes end up professing the most ab-
surd views. What we must explain are the patterns: why some
ideas, including good ones, are so hard to get across, while others,
including bad ones, are so popu lar.
Mechanisms of Open Vigilance
Understanding our mechanisms of open vigilance is the key to
making sense of the successes and failures of communication.
These mechanisms pro cess a variety of cues to tell us how much
we should believe what we’re told. Some mechanisms examine
whether a message is compatible with what we already believe
to be true, and whether it is supported by good arguments. Other
mechanisms pay attention to the source of the message: Is the
speaker likely to have reliable information? Does she have my
interests at heart? Can I hold her accountable if she proves
mistaken?
I review a wealth of evidence from experimental psy chol ogy
showing how well our mechanisms of open vigilance function,
including in small children and babies. It is thanks to these mech-
anisms that we reject most harmful claims. But these mecha-
nisms also explain why we accept a few mistaken ideas.
For all their sophistication, and their capacity to learn and in-
corporate novel information, our mechanisms of open vigilance
in t r o d u c t i o n xvii
are not infinitely malleable. You, dear reader, are in an informa-
tion environment that differs in myriad ways from the one your
ancestors evolved in. You are interested in people you’ll never
meet (politicians, celebrities), events that don’t affect you (a di-
saster in a distant country, the latest scientific breakthrough),
and places you’ll never visit (the bottom of the ocean, galaxies
far, far away). You receive much information with no idea of
where it came from: Who started the rumor that Elvis wasn’t
dead? What is the source of your parents’ religious beliefs? You
are asked to pass judgment on views that had no practical rele-
vance whatsoever for our ancestors: What is the shape of the
earth? How did life evolve? What is the best way to or ga nize a
large economic system? It would be surprising indeed if our
mechanisms of open vigilance functioned impeccably in this
brave new, and decidedly bizarre, world.
Our current informational environment pushes open vigi-
lance mechanisms outside of their comfort zone, leading to
mistakes. On the whole, we are more likely to reject valuable
messages— from the real ity of climate change to the efficacy of
vaccination— than to accept inaccurate ones. The main excep-
tions to this pattern stem not so much from a failure of open vigi-
lance itself, but from issues with the material it draws on. People
sensibly use their own knowledge, beliefs, and intuitions to evalu-
ate what they’re told. Unfortunately, in some domains our in-
tuitions appear to be quite systematically mistaken. If you had
nothing else to go on, and someone told you that you were stand-
ing on a flat surface (rather than, say, a globe), you would spon-
taneously believe them. If you had nothing else to go on, and
someone told you all your ancestors had always looked pretty
much like you (and not like, say, fish), you would spontaneously
believe them. Many popu lar yet mistaken beliefs spread not
xviii in t roduc t ion
because they are pushed by masters of persuasion but because
they are fundamentally intuitive.
If the flatness of the earth is intuitive, a two- hundred- foot-
high, thousands- of- miles- long wall of ice is not. Nor is, say, Kim
Jong- il’s ability to teleport. Reassuringly, the most out- there be-
liefs out there are accepted only nominally. I bet a flat- earther
would be shocked to actually run into that two- hundred- foot
wall of ice at the end of the ocean. Seeing Kim Jong-il being
beamed Star Trek– style would have confused the hell out of the
dictator’s most groveling sycophant. The critical question for un-
derstanding why such beliefs spread is not why people accept
them, but why people profess them. Besides wanting to share
what we take to be accurate views, there are many reasons for
professing beliefs: to impress, annoy, please, seduce, manipulate,
reassure. These goals are sometimes best served by making state-
ments whose relation to real ity is less than straightforward—or
even, in some cases, statements diametrically opposed to the